Altered Expression of MicroRNA-15a and Kruppel-Like Factor 4 in Cerebrospinal Fluid and Plasma After Aneurysmal Subarachnoid Hemorrhage

Accepted Manuscript Altered Expression of microRNA-15a and Kruppel-like Factor 4 in Cerebrospinal Fluid and Plasma after Aneurysmal Subarachnoid Hemorrhage Yuichiro Kikkawa, Takeshi Ogura, Hiroyuki Nakajima, Toshiki Ikeda, Ririko Takeda, Hiroaki Neki, Shinya Kohyama, Fumitaka Yamane, Ryota Kurogi, Toshiyuki Amano, Akira Nakamizo, Masahiro Mizoguchi, Hiroki Kurita PII:

S1878-8750(17)31512-7

DOI:

10.1016/j.wneu.2017.09.008

Reference:

WNEU 6452

To appear in:

World Neurosurgery

Received Date: 2 June 2017 Revised Date:

1 September 2017

Accepted Date: 2 September 2017

Please cite this article as: Kikkawa Y, Ogura T, Nakajima H, Ikeda T, Takeda R, Neki H, Kohyama S, Yamane F, Kurogi R, Amano T, Nakamizo A, Mizoguchi M, Kurita H, Altered Expression of microRNA-15a and Kruppel-like Factor 4 in Cerebrospinal Fluid and Plasma after Aneurysmal Subarachnoid Hemorrhage, World Neurosurgery (2017), doi: 10.1016/j.wneu.2017.09.008. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Kikkawa et al.

ACCEPTED MANUSCRIPT

Altered Expression of microRNA-15a and Kruppel-like Factor 4 in Cerebrospinal Fluid and Plasma after Aneurysmal Subarachnoid Hemorrhage

RI PT

Yuichiro Kikkawa1, Takeshi Ogura1, Hiroyuki Nakajima1, Toshiki Ikeda1, Ririko Takeda1, Hiroaki Neki2, Shinya Kohyama2, Fumitaka Yamane2, Ryota Kurogi3, Toshiyuki Amano3,

SC

Akira Nakamizo3, Masahiro Mizoguchi4, Hiroki Kurita1

1) Department of Cerebrovascular Surgery, Saitama Medical University International Medical

M AN U

Center, Hidaka, Japan

2) Department of Endovascular Neurosurgery, Saitama Medical University International Medical Center, Hidaka, Japan

3) Department of Neurosurgery, Clinical Research Institute, National Hospital Organization

TE D

Kyushu Medical Center, Fukuoka, Japan

4) Department of Neurosurgery, Kitakyushu Municipal Medical Center, Kitakyushu, Japan.

EP

Correspondence: Yuichiro Kikkawa, M.D., Ph.D.

Department of Cerebrovascular Surgery, Saitama Medical University International Medical Center,

AC C

1397-1 Yamane, Hidaka, Saitama 350-1298, Japan Fax: +81-42-984-4304, tel: +81-42-984-4423 E-mail: [email protected]

Key words: subarachnoid hemorrhage, microRNAs, cerebral vasospasm, vascular phenotype

Abbreviations list: CVS: cerebral vasospasm SAH: subarachnoid hemorrhage R1

Kikkawa et al.

ACCEPTED MANUSCRIPT

miRNAs: microRNAs miR-15a : microRNA-15a KLF4: Kruppel-like factor 4 CSF: cerebrospinal fluid

RI PT

VSMCs: vascular smooth muscle cells

AC C

EP

TE D

M AN U

SC

ECs: endothelial cells

R2

Kikkawa et al.

ACCEPTED MANUSCRIPT

Abstract Background: Cerebral vasospasm (CVS) is a major determinant of prognosis in patients with subarachnoid hemorrhage (SAH). Alteration in the vascular phenotype contributes to development

RI PT

of CVS. However, little is known about the role of microRNAs (miRNAs) in the phenotypic

alteration after SAH. We investigated the expression profile of miRNAs and the chronological

changes in the expression of microRNA-15a (miR-15a) and Kruppel-like factor 4 (KLF4), a potent

SC

regulator of vascular phenotype modulation that modulates the expression of miR-15a, in the plasma and cerebrospinal fluid (CSF) of SAH patients.

M AN U

Methods: Peripheral blood and CSF samples were collected from eight aneurysmal SAH patients treated with endovascular obliteration. Samples obtained from three patients without SAH were used as controls in the analysis. Exosomal miRNAs were isolated and subjected to microarray analysis with the 3D-gene miRNA microarray kit. The time course of the expression of miR-15a and KLF4

TE D

was analyzed using quantitative real-time PCR.

Result: Microarray analysis demonstrated that 12 miRNAs including miR-15a were up- or down-regulated both in the CSF and plasma after SAH within three days. Quantitative real-time

EP

PCR revealed that miR-15a expression was significantly increased in both the CSF and plasma, with a peak around 3 to 5 days after SAH, whereas the expression of KLF4 was significantly decreased

AC C

around 1 to 3 days after SAH and remained lower than in controls. Conclusion: Our results suggest that an early and persistent decrease in KLF4 followed by an increase in miR-15a may contribute to the altered vascular phenotype, resulting in development of CVS.

R3

Kikkawa et al.

ACCEPTED MANUSCRIPT

Introduction Cerebral vasospasm (CVS) is one of the most important cerebrovascular events following aneurysmal subarachnoid hemorrhage (SAH) and is characterized by delayed and prolonged

RI PT

narrowing of cerebral arteries that may cause delayed cerebral ischemia (DCI), leading to death or neurological deficits in patients with SAH 1. Alterations in the vascular phenotype contribute

substantially to development of CVS 2-5. Phenotypic alteration of vascular smooth muscle cells

SC

(VSMCs) causes increased vascular contractility through various vasoconstriction mechanisms such as up-regulation of receptor activity and increased myofilament Ca2+ sensitization of the contractile

M AN U

apparatus following SAH 6-8. Moreover, functional reduction in vasodilatory mechanisms such as an endothelial dysfunction may also contribute to an increased vascular contractility following SAH 9. On the other hand, vascular phenotypic alterations are important in the regulation of vessel wall compliance following SAH. Vasospasm of arteries is characterized by intimal thickening and wall

TE D

stiffness, which is induced by vascular cell proliferation and extracellular matrix metabolism, contributing to the maintenance of vasospasm 2,4,5.

MicroRNAs (miRNAs) inhibit gene expression at a posttranscriptional level by interfering

EP

with translational initiation or degradation of mRNA and play key regulatory roles in a wide range of genetic pathways involved in many biological processes 10. Recent studies suggest that alterations in

AC C

VSMC phenotypes are regulated by miRNAs in various pathological conditions including myocardial infarction, aneurysms, arteriosclerosis, hypertension, and respiratory diseases 11,12. miR-15a, which was discovered in chronic lymphocytic leukemia, acts as a tumor suppressor or a potential oncogenic miRNA in many types of cancer 13. Recent studies have revealed that miR-15a is involved in vascular angiogenesis or proliferation of endothelial cells (ECs) and VSMCs in ischemia-induced cerebral vascular endothelial injury or hindlimb ischemia 14,15. Furthermore, miR-15a is also closely involved in cerebrovascular protection following ischemia 16. However, little

R4

Kikkawa et al.

ACCEPTED MANUSCRIPT

is known about the role of miRNAs in the pathogenesis of CVS following SAH 17-21. Kruppel-like factor 4 (KLF4) includes a zinc finger structure and is a member of the Kruppel-like factor family of transcription factors. KLF4 is highly expressed in ECs and VSMCs

RI PT

including these cells in the cerebral vasculature 22-24. Accumulating evidence has shown that KLF4 contributes to the regulation of phenotypic modulation of VSMCs in various disease states including vascular injury, atherosclerosis, hemodynamic stress, and cerebral aneurysm 22,25-27. However, the

SC

role of KLF4 in the development of CVS is unknown. Recently, the relationship between KLF4 and miR-15a was partially revealed. KLF4 inhibits the proliferation and angiogenesis of ECs and

M AN U

VSMCs through the induction of miR-15a 28.

The aim of this study was to elucidate the underlying mechanism involved in the vascular phenotypic alteration following SAH. In the present study, microarray analysis of exosomal miRNAs was conducted and quantitative real-time PCR analysis was used to investigate the change

AC C

EP

patients over time.

TE D

in the expression levels of miR-15a and KLF4 in the plasma and cerebrospinal fluid (CSF) in SAH

R5

Kikkawa et al.

ACCEPTED MANUSCRIPT

Materials and Methods Patient recruitment and sample collection This study was performed in accordance with the Japanese government’s Ethical Guidelines for

RI PT

Medical and Health Research Involving Human Subjects (2014) and the declaration of Helsinki, and was approved by the institutional review board (IRB) of Saitama Medical University International Medical Center (IRB No. 14-099). Written informed consent was obtained for the collection of

SC

samples and subsequent analysis of all participants. Eight SAH patients who underwent

endovascular obliteration of cerebral aneurysms at Saitama Medical University International

M AN U

Medical Center within 24 hours of onset were enrolled in this study. The ages of the patients ranged from 36 to 84 years (mean age 64 years). Five patients were female. The preoperative neurological status was classified according to the Hunt and Kosnik grading scale (Grade II: 4, III: 3, IV: 1). The distribution and amount of subarachnoid blood clots were assessed with the Fisher scale 29, and all

TE D

patients were classified into Fisher group 3. Five aneurysms were located in the anterior circulation, and three were in the posterior circulation. In all cases, complete obliteration of the aneurysm was achieved by simple coiling without any intraprocedural vascular injuries, such as aneurysmal

EP

perforation or arterial dissection. Postoperatively, cilostazol and fasudil hydrochloride were used as prophylaxis for CVS. In this study, we did not use the Ca-channel blocker nimodipine, which is

AC C

widely used as a prophylactic agent against delayed ischemic neurological deficits following SAH worldwide, because it has not been approved in Japan. However, in Japan, various other prophylactics, such as fasudil hydrochloride and ozagrel sodium, which are highly recommended in Japanese guidelines, are usually used in many hospitals. The presence of vasospasm was evaluated using CT angiography at approximately day 7 after SAH. As a result, angiographic vasospasm was not observed in any patients with SAH. In all patients, the postoperative course was uneventful, and no patient developed delayed ischemic neurological deficits, delayed cerebral infarction, or central

R6

Kikkawa et al.

ACCEPTED MANUSCRIPT

nervous system infection. On day 1, a lumbar catheter was placed through the lumbar vertebral interspace into the subarachnoid space, and external lumbar drainage of CSF was continuously performed for 10 days.

RI PT

CSF (4 mL) samples were collected from the lumbar catheter into a 10-mL sterile tube on days 1, 3, 5, 7, 9, and 11.

On days 1, 3, 5, 7, 9, 11, and 13, peripheral blood samples (3 mL) were collected into 5-mL

SC

EDTA-containing tubes. As controls (day 0), CSF and blood samples were obtained from three

patients (mean age 66 years) with suspected idiopathic normal pressure hydrocephalus, although a

M AN U

CSF tap test was negative for idiopathic normal pressure hydrocephalus in all three patients. These three patients also had no major medical problems including a history of smoking or hypertension. Whole blood and CSF samples were immediately centrifuged at 12000 rpm for 15 min at 4°C. The plasma fraction of blood and the CSF lacking blood cell components were aliquoted into 1.5-mL

shown in Figure 1A.

TE D

microcentrifuge tubes and stored at −80°C until subsequent assays. The timeline of sampling is

EP

Microarray analysis of exosomal miRNAs

The blood and CSF samples collected on day 3 from two SAH patients and control samples were

AC C

used for microarray analysis. Exosomes were isolated from 500 µL plasma and 1 mL CSF using the ExoQuick Plasma prep and Exosome precipitation kit (System Biosciences Inc., Mountain View, CA, USA) and ExoQuick-TC precipitation solution (System Biosciences Inc.), respectively, according to the manufacturer’s instructions. Total RNA was isolated from exosome samples using the total exosome RNA and protein isolation kit (Invitrogen, Carlsbad, CA, USA). RNA integrity and the presence of the small RNA fraction were confirmed using a Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA), and the concentration was determined using NanoDrop ND-1000 and -2000

R7

Kikkawa et al.

ACCEPTED MANUSCRIPT

(NanoDrop Technologies, Wilmington, DE, USA). Microarray gene expression analysis was performed using the 3D-gene miRNA microarray platform (V21_1.0.0; Toray Industries, Kamakura, Japan), which detects 2,565 human miRNAs. Briefly, total RNA was labeled and hybridized using a

RI PT

3D-gene miRNA labeling kit (Toray Industries) for 16 hours at 32°C. The microarray was washed, and hybridization signals were detected using a 3D-Gene Scanner 3000 (Toray Industries). Fluorescent signals were extracted from the scanned images to generate raw data using 3D-Gene

SC

extraction software (Toray Industries). The raw data for each spot were normalized by subtraction of the mean intensity of the background signal determined from the signal intensities of all blank spots

M AN U

with 95% confidence intervals. The signal intensities of raw data greater than mean + 2 standard deviations (SD) of the background signal intensity were considered to be valid. Finally, detected signals for each gene were globally normalized per microarray by adjusting the median of the signal intensity to 25. Changes in the level of signal intensity over 50 were considered to be significant in

TE D

all the arrays, and signal intensities were log2 transformed. The signal level of the miRNAs on day 3 with log2 fold-change (|log2 fold change|) higher than 1.0 compared to the control (day 0) were defined as up-regulated in this study, whereas the signal level of the miRNAs on day 3 with log2

EP

fold-change less than −1.0 compared to the control (day 0) were defined as down-regulated. The expression level of miRNAs on day 3 after SAH was represented as the log2 fold change of the

AC C

control (day 0). Microarray data are available from the Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/) with the accession number GSE93746.

Quantitative real-time polymerase chain reaction (PCR) analysis of expression of exosomal miRNAs Exosomal miRNAs were reverse transcribed with the Light Cycler 480 System (Roche Diagnostics, Basel, Switzerland) using the TaqMan microRNA reverse transcription kit (#4366596, Applied

R8

Kikkawa et al.

ACCEPTED MANUSCRIPT

Biosystems, Foster City, CA, USA) and the associated miRNA-specific primers in a 15-µL reaction mixture containing 0.15 µL dNTP (100 mM), 1 µL MultiScribe reverse transcriptase (50 U/µL; Applied Biosystems), 1.5 µL 10× RT buffer, and 0.19 µL RNase inhibitor (20 U/µL). The reaction

RI PT

conditions were as follows: 16°C for 30 min, 42°C for 30 min, and 85°C for 5 min. For miR-15a, the miRNA-specific primer from the TaqMan microRNA assay kit (#4427975; Applied Biosystems) was used. For microRNA-6724 (miR-6724), custom small RNA TaqMan assays were designed and

SC

synthesized by Applied Biosystems.

Real-time PCR was performed in triplicate using the LightCycler 480 System II (Roche

M AN U

Diagnostics). PCR was performed using the TaqMan small RNA assay kit (Applied Biosystems) with an initial denaturing step at 50°C for 2 min and 95°C for 10 min, followed by 50 cycles of 95°C for 15 s and 60°C for 1 min. Each PCR contained 5 µL LightCycler 480 probes master (Roche Diagnostics), 2 µL cDNA, and 0.5 µL TaqMan small RNA assay reagent (Applied Biosystems) in a

TE D

total volume of 10 µL. The relative microRNA levels were normalized to endogenous microRNA-4655 (miR-4655) expression for each sample. miR-4655 was used as an internal standard in quantitative real-time PCR experiments of miRNAs, because it was detectable in all

EP

samples and its normalized intensities were not significantly different between control and SAH samples in either serum or CSF (ratio was between 0.8 and 1.25 in the comparison of all pairs of

AC C

samples) in the microarray analysis. The relative levels of gene expression were represented as ∆Ct = Ct gene – Ct reference, and the fold change in gene expression was calculated by the 2−∆∆Ct method 30. The levels of miRNA expression on day 0 were assigned a value of 1.0. The data were expressed as the mean values ± SEM.

Quantitative real-time PCR analysis of KLF4 Total RNA was isolated from 0.3 mL plasma and 0.3 mL CSF using the Maxwell 16 LEV

R9

Kikkawa et al.

ACCEPTED MANUSCRIPT

simplyRNA cells kit (Promega, Madison, WI, USA), according to the manufacturer's protocols. Total mRNAs were reverse transcribed into cDNAs using the ReverTra Ace qPCR RT kit (Toyobo, Osaka, Japan) in a 10-µL reaction mixture containing 0.5 µL reverse transcriptase, 2 µL 10× RT buffer, and

RI PT

0.5 µL primer mixture reagent with the LightCycler 480 System (Roche Diagnostics). The reaction conditions were as follows: 37°C for 15 min and 98°C for 2 min. Real-time PCR was performed in triplicate using the LightCycler 480 System II (Roche Diagnostics). PCR was performed using the

SC

TaqMan gene expression assay (#4331182; Applied Biosystems) with an initial denaturing step at 50°C for 2 min and 95°C for 10 min, followed by 50 cycles of 95°C for 15 s and 60°C for 1 min.

M AN U

Each PCR contained 5 µL LightCycler 480 probes master (Roche Diagnostics), 2 µL cDNA, and 0.5 µL TaqMan gene expression assay mixes (Applied Biosystems) for KLF4 in a total volume of 10 µL. The relative mRNA levels were normalized to β-actin expression for each sample and calculated as described above. The levels of KLF4 expression on day 0 were assigned a value of 1.0. The data

TE D

were expressed as the mean values ± SEM. The following human DNA-specific primer pairs were used for KLF4 (NM_004235); forward (5′–3′) CTGCGAACCCACACAGGTG, reverse (5′–3′)

Data analysis

EP

GGTAGTGCCTGGTCAGTTCATC. The investigation flowchart is shown in Figure 1B.

AC C

The PCR data are expressed as the mean ± SEM of the indicated experimental number. Analysis of variance (ANOVA) followed by Dunnett’s test was performed to determine statistically significant differences in a multiple comparison with the control data. A value of P < 0.05 was considered to be statistically significant, unless otherwise specified. All analyses were performed using GraphPad PRISM software version 6.0 (GraphPad software, San Diego, CA, USA).

R10

Kikkawa et al.

ACCEPTED MANUSCRIPT

Results Microarray analysis of plasma and CSF miRNA following SAH Because gene expression changes dramatically during the early phase after SAH 31, we investigated

RI PT

the changes in miRNA expression on day 3 after SAH using microarray analysis. Among the investigated 2,565 miRNAs, 27 miRNAs were up-regulated and 13 miRNAs were down-regulated in the CSF, whereas four miRNAs were up-regulated and 15 miRNAs were down-regulated in the

SC

plasma in two SAH samples on day 3 after SAH (Supplementary Table S1). Among these miRNAs, 12 miRNAs were up- or down-regulated in both the CSF and plasma in two SAH samples (Table 1),

M AN U

suggesting that these miRNAs may be important in development of post-SAH pathological condition. Accordingly, these 12 miRNAs were investigated further.

Time course of changes in the expression of miR-15a in the plasma and CSF following SAH

TE D

Among the 12 miRNAs that were up- or down-regulated in both CSF and plasma, we focused on miR-15a, which is involved in vascular angiogenesis and proliferation implicated in cerebrovascular protection after stroke

14,15,28

and is closely

16

. Before evaluation of miR-15a, to verify

EP

whether the results from the microarray analysis reflect early changes in miRNA expression after SAH, quantitative real-time PCR for two differentially expressed miRNAs, miR-15a and miR-6724,

AC C

was performed using samples from days 1 and 3. As a result, the expression level of miR-15a was decreased from day 1 to day 3 in the plasma, whereas the expression level of miR-15a was increased from day 1 to day 3 in the CSF (Supplementary Fig. S1 A, B). On the other hand, the expression level of miR-6724 was decreased from day 1 to day 3 in both the plasma and CSF (Supplementary Fig. S1 C, D). These results from real-time PCR were consistent with those from microarray analysis. Next, we investigated the time course of changes in the expression of miR-15a in the plasma

R11

Kikkawa et al.

ACCEPTED MANUSCRIPT

and CSF. Quantitative real-time PCR revealed that miR-15a was significantly up-regulated on days 5 and 7 with a peak on day 5 in the plasma (Fig. 2A) and was significantly up-regulated on days 1, 3,

RI PT

and 5 with a peak on day 3 in the CSF (Fig. 2B).

Time course of changes in the mRNA expression of KLF4 in the plasma and CSF following SAH

SC

KLF4 suppresses the proliferation and angiogenesis of ECs and VSMCs by inducing miR-15a

28

.

Therefore, the time course of SAH-induced changes in KLF4 mRNA expression was analyzed using

M AN U

quantitative real-time PCR in the plasma and CSF. In the plasma, KLF4 mRNA expression was significantly decreased on day 3 and remained significantly lower than in the control throughout the time course (Fig. 3A). On the other hand, in the CSF, KLF4 mRNA expression was significantly decreased on day 1, remained significantly low until day 9, and then was restored to the control level

Discussion

TE D

on day 11 (Fig. 3B).

EP

The present study showed the time course of changes in the expression of miR-15a and KLF4 in the CSF and blood after SAH. Based on microarray analysis, we identified miR-15a as a miRNA that

AC C

may be involved in the alteration of the vascular phenotype following SAH. Quantitative real-time PCR revealed that KLF4 expression was significantly decreased both in the CSF and plasma early after SAH. Subsequently, miR15a expression was significantly increased, with a peak around 3 to 5 days after SAH. KLF4 expression was significantly decreased early after SAH and remained lower than in the control in both the CSF and plasma. This suggests that production of KLF4 is continuously reduced in some tissues after the onset of SAH. KLF4 is highly expressed in various tissues including ECs

R12

Kikkawa et al.

ACCEPTED MANUSCRIPT

and VSMCs of the cerebral vasculature and acts as a critical repressor of vascular cell proliferation 24,28

, contributing to regulation of vascular phenotypic modulation in various diseases.

Overexpression of KLF4 inhibits neointimal hyperplasia induced by balloon injury by inhibiting 32

. In contrast, conditional knockout of KLF4 accelerates injury-induced cellular

proliferation of VSMCs in mice

33

RI PT

c-Jun expression

. In the present study, KLF4 expression was significantly and

continuously decreased in the early period after SAH. Accordingly, if the decreased expression level

SC

of KLF4 in the plasma and CSF is attributable to decreased expression in ECs or VSMCs of cerebral vessels, an early and persistent decrease in KLF4 in cerebral vessels may contribute to enhancement

M AN U

of cerebral vascular proliferation, leading to development of delayed CVS

2,4,5

. However, the

mechanisms underlying the early reduction in KLF4 remain to be elucidated. The expression level of miR-15a was increased in the CSF and plasma with a peak on days 3 and 5, respectively. Little is known about the role of miRNAs in the pathogenesis of post-SAH

TE D

events including early brain injury, CVS, or DCI. Several studies have investigated the alteration in miRNA expression profiles after SAH using microarray analysis in humans or animal models

17-21

.

Among them, two studies showed that one of the miRNAs that was upregulated in CSF or blood was 17,19

. Our results are consistent with these preliminary reports. In the present study, we

EP

miR-15a

further demonstrated the details of the chronological changes in miR-15a expression and clearly

AC C

showed elevation of miR-15a using real-time PCR in both the CSF and plasma. However, the roles and meaning of miR-15a expression in the development of post-SAH pathology remain unknown. In this regard, we propose two possible roles for miR-15a expression following SAH. One possible role for miR-15a is related to cerebral ischemia. miR-15a is closely implicated in cerebrovascular protection following ischemic stroke vascular injury

14

16

and pathogenesis of ischemic

. Gain or loss of miR-15a significantly reduces or increases oxygen-glucose

deprivation-induced cerebral vascular endothelial cell death, respectively 14. Furthermore, circulating

R13

Kikkawa et al.

ACCEPTED MANUSCRIPT

miR-15a is a potential biomarker after ischemic stroke in both animal models and humans

34,35

. In

the present study, plasma miR-15a peaked on day 5 and remained at a high level on day 7 after SAH. Generally, DCI, which is a major cause of clinical deterioration of SAH patients, occurs mostly

RI PT

between days 4 and 10 after onset of SAH 1. The timing of the peak of plasma miR-15a elevation in the present study is consistent with this period. Therefore, the up-regulation of miR-15a may at least partially represent a molecular mechanism that causes DCI after SAH. In the present study,

SC

quantitative real-time PCR also demonstrated a small but significant increase in miR-15a expression in the plasma on day 1. If the elevation of miR-15a reflects cerebral ischemia, this elevation may

M AN U

also reflect the insult caused by transient global ischemia, the fundamental pathology causing early brain injury following SAH, just after the onset of SAH.

Another possible role for miR-15a is related to vascular angiogenesis or proliferation

19,20,28

.

miR-15a negatively regulates angiogenesis by suppression of fibroblast growth factor 2 and vascular 20

. Intriguingly, in the present study, KLF4

TE D

endothelial growth factor in vascular endothelial cells

was dramatically decreased prior to the peak of miR-15a elevation both in the plasma and CSF. Little is known about the relationship between KLF4 and miR-15a 28. Zheng et al. demonstrated that KLF4

EP

inhibits the proliferation of ECs and VSMCs, as well as angiogenesis, via induction of miR-15a 28. They revealed that KLF4 increase the expression of miR-15 in ECs and VSMCs, and functional

AC C

inhibition of miR-15a largely alleviates the suppressive effect of KLF4 on the proliferations of ECs and VSMCs, as well as angiogenesis

28

. Therefore, our present data showing early and persistent

down-regulation of KLF4 suggest that anti-proliferative and anti-angiogenic effects mediated by KLF4-induced miR-15a up-regulation were sustainably attenuated from the early period after SAH, contributing to development of CVS through vascular proliferation. However, in the present study, miR-15a underwent a transient increase after a decrease in KLF4, rather than a corresponding decrease, suggesting that the elevation of miR-15a is caused by some mechanisms independently

R14

Kikkawa et al.

ACCEPTED MANUSCRIPT

from KLF4-mediated miR-15a up-regulation. Therefore, here, we speculate that miR-15a is up-regulated to compensate for the diminished anti-proliferative and anti-angiogenic effects of KLF4 following a decrease in KLF4, a situation that may represent a feedback regulation mechanism for

However, further studies are necessary to fully explore these ideas.

RI PT

impaired miR-15a-mediated anti-proliferative and anti-angiogenic effects of KLF4 following SAH.

Generally, miRNAs are thought to be released from damaged cells or circulating cells, 36

. Therefore, miR-15a expression in blood is

SC

leading to an increase in blood miRNA expression

likely derived from damaged tissues including ECs and VSMCs after SAH. In the present study, the

M AN U

peak of miR-15a expression in the CSF occurred prior to the peak in the plasma. This time delay between peaks in miR-15a in the CSF and plasma may reflect the difference in timing in which miR-15a is secreted into CSF and blood from damaged tissues. The first miR-15a peak in the CSF may be due to a release from cells through the damaged adventitia, probably caused by initial blood 37,38

. Taken

TE D

extravasation into the subarachnoid space and subsequent blood-brain barrier disruption

together, the second miR-15a peak in the plasma may be induced by the secretion of miR-15a from ECs or VSMCs via the disrupted endothelium 9. Further investigations using animal models are

EP

needed to confirm the localization of miR-15a expression following SAH. This study has a limitation regarding the sample size. Only eight patients were included in this

AC C

study, and all of them had an uneventful postoperative course without DCI or CVS. Thus, we did not analyze the potential for these molecules to be used as prognostic biomarkers for DCI or CVS. A larger sample size will provide further insight into the role of these molecules and their potential as biomarkers for prognosis of DCI or CVS. Furthermore, the pattern of changes in the expression of these molecules under interaction of various drugs should be investigated to determine whether these molecules could be therapeutic targets. On the other hand, a strength of this study is the uniformity in samples. All patients in this study were treated with endovascular coiling. Therefore, the

R15

Kikkawa et al.

ACCEPTED MANUSCRIPT

alterations in molecular expression in the CSF or blood due to treatment of the aneurysm are much smaller than alterations following clipping surgery, which requires a craniotomy and intracranial manipulations. Furthermore, the blood distribution and amount of blood clots of all patients were in

RI PT

Fisher group 3. Thus, our data are reliable regarding the expression analysis of the CSF.

Conclusion

SC

The present study showed the time course of changes in the expression of miR-15a and KLF4 in the CSF and plasma following SAH. Our results suggest that an early and persistent decrease in KLF4

M AN U

followed by an increase in miR-15a may contribute to vascular proliferation or angiogenesis, resulting in CVS. Further investigations using large numbers of samples and animal models are needed to clarify the role of these molecules in the development of CVS.

TE D

Acknowledgement

This research was supported by JSPS KAKENHI Grant numbers 26462164, 25670624 and 17K10848.

EP

We thank Ms. Kozue Watanabe for technical assistance.

AC C

Conflicts of interest

The authors declare that there is no conflict of interest regarding the publication of this article.

R16

Kikkawa et al.

ACCEPTED MANUSCRIPT

References Kassell NF, Sasaki T, Colohan AR, Nazar G. Cerebral vasospasm following aneurysmal subarachnoid hemorrhage. Stroke. 1985;16:562-572. 2.

RI PT

1.

Borel CO, McKee A, Parra A, Haglund MM, Solan A, Prabhakar V, et al. Possible role for

SC

vascular cell proliferation in cerebral vasospasm after subarachnoid hemorrhage. Stroke. 2003;34:427-433.

Edvinsson LI, Povlsen GK. Vascular plasticity in cerebrovascular disorders. J Cereb Blood Flow Metab. 2011;31:1554-1571.

4.

M AN U

3.

Kurogi R, Kikkawa Y, Matsuo S, Nakamizo A, Mizoguchi M, Sasaki T. Upregulation of tissue inhibitor of metalloproteinase-1 contributes to restoration of the extracellular matrix

TE D

in the rabbit basilar artery during cerebral vasospasm after subarachnoid hemorrhage. Brain Res. 2015;1616:26-36. 5.

Zhou C, Yamaguchi M, Colohan AR, Zhang JH. Role of p53 and apoptosis in cerebral

EP

vasospasm after experimental subarachnoid hemorrhage. J Cereb Blood Flow Metab. 2005;25:572-582.

Hansen-Schwartz J, Ansar S, Edvinsson L. Cerebral vasoconstriction after subarachnoid

AC C

6.

hemorrhage--role of changes in vascular receptor phenotype. Front Biosci. 2008;13:2160-2164.

7.

Kikkawa Y, Kameda K, Hirano M, Sasaki T, Hirano K. Impaired feedback regulation of the receptor activity and the myofilament Ca2+ sensitivity contributes to increased vascular reactiveness after subarachnoid hemorrhage. J Cereb Blood Flow Metab. 2010;30:1637-1650.

R17

Kikkawa et al.

ACCEPTED MANUSCRIPT

8.

Kikkawa Y, Matsuo S, Kameda K, Hirano M, Nakamizo A, Sasaki T, et al. Mechanisms underlying potentiation of endothelin-1-induced myofilament Ca(2+) sensitization after subarachnoid hemorrhage. J Cereb Blood Flow Metab. 2012;32:341-352. Sasaki T, Kassell NF. The role of endothelium in cerebral vasospasm. Neurosurg Clin N Am.

RI PT

9.

1990;1:451-463. 10.

Small EM, Olson EN. Pervasive roles of microRNAs in cardiovascular biology. Nature.

11.

SC

2011;469:336-342.

Mohamed JS, Lopez MA, Boriek AM. Mechanical stretch up-regulates microRNA-26a and

M AN U

induces human airway smooth muscle hypertrophy by suppressing glycogen synthase kinase-3beta. J Biol Chem. 2010;285:29336-29347. 12.

Santulli G. microRNAs Distinctively Regulate Vascular Smooth Muscle and Endothelial Cells: Functional Implications in Angiogenesis, Atherosclerosis, and In-Stent Restenosis.

TE D

Adv Exp Med Biol. 2015;887:53-77. 13.

Pekarsky Y, Croce CM. Role of miR-15/16 in CLL. Cell Death Differ. 2015;22:6-11.

14.

Yin KJ, Deng Z, Hamblin M, Xiang Y, Huang H, Zhang J, et al. Peroxisome

EP

proliferator-activated receptor delta regulation of miR-15a in ischemia-induced cerebral vascular endothelial injury. J Neurosci. 2010;30:6398-6408. Yin KJ, Olsen K, Hamblin M, Zhang J, Schwendeman SP, Chen YE. Vascular endothelial

AC C

15.

cell-specific microRNA-15a inhibits angiogenesis in hindlimb ischemia. J Biol Chem.

2012;287:27055-27064.

16.

Yin KJ, Fan Y, Hamblin M, Zhang J, Zhu T, Li S, et al. KLF11 mediates PPARgamma cerebrovascular protection in ischaemic stroke. Brain. 2013;136:1274-1287.

17.

Stylli SS, Adamides AA, Koldej RM, Luwor RB, Ritchie DS, Ziogas J, et al. miRNA expression profiling of cerebrospinal fluid in patients with aneurysmal subarachnoid

R18

Kikkawa et al.

ACCEPTED MANUSCRIPT

hemorrhage. J Neurosurg. 2016:1-9. 18.

Powers CJ, Dickerson R, Zhang SW, Rink C, Roy S, Sen CK. Human cerebrospinal fluid microRNA: temporal changes following subarachnoid hemorrhage. Physiol Genomics.

19.

RI PT

2016;48:361-366.

Su XW, Chan AH, Lu G, Lin M, Sze J, Zhou JY, et al. Circulating microRNA 132-3p and 324-3p Profiles in Patients after Acute Aneurysmal Subarachnoid Hemorrhage. PLoS One.

20.

SC

2015;10:e0144724.

Bache S, Rasmussen R, Rossing M, Hammer NR, Juhler M, Friis-Hansen L, et al. Detection

21.

M AN U

and quantification of microRNA in cerebral microdialysate. J Transl Med. 2015;13:149. Muller AH, Povlsen GK, Bang-Berthelsen CH, Kruse LS, Nielsen J, Warfvinge K, et al. Regulation of microRNAs miR-30a and miR-143 in cerebral vasculature after experimental subarachnoid hemorrhage in rats. BMC Genomics. 2015;16:119. Ali MS, Starke RM, Jabbour PM, Tjoumakaris SI, Gonzalez LF, Rosenwasser RH, et al.

TE D

22.

TNF-alpha induces phenotypic modulation in cerebral vascular smooth muscle cells: implications for cerebral aneurysm pathology. J Cereb Blood Flow Metab.

23.

EP

2013;33:1564-1573.

Deaton RA, Gan Q, Owens GK. Sp1-dependent activation of KLF4 is required for

AC C

PDGF-BB-induced phenotypic modulation of smooth muscle. Am J Physiol Heart Circ Physiol. 2009;296:H1027-1037.

24.

Zheng B, Han M, Bernier M, Zhang XH, Meng F, Miao SB, et al. Kruppel-like factor 4

inhibits proliferation by platelet-derived growth factor receptor beta-mediated, not by retinoic acid receptor alpha-mediated, phosphatidylinositol 3-kinase and ERK signaling in vascular smooth muscle cells. J Biol Chem. 2009;284:22773-22785. 25.

Regan CP, Adam PJ, Madsen CS, Owens GK. Molecular mechanisms of decreased smooth

R19

Kikkawa et al.

ACCEPTED MANUSCRIPT

muscle differentiation marker expression after vascular injury. J Clin Invest. 2000;106:1139-1147. 26.

Shi ZD, Abraham G, Tarbell JM. Shear stress modulation of smooth muscle cell marker

and ERK1/2. PLoS One. 2010;5:e12196. 27.

RI PT

genes in 2-D and 3-D depends on mechanotransduction by heparan sulfate proteoglycans

Wamhoff BR, Hoofnagle MH, Burns A, Sinha S, McDonald OG, Owens GK. A G/C

SC

element mediates repression of the SM22alpha promoter within phenotypically modulated smooth muscle cells in experimental atherosclerosis. Circ Res. 2004;95:981-988. Zheng X, Li A, Zhao L, Zhou T, Shen Q, Cui Q, et al. Key role of microRNA-15a in the

M AN U

28.

KLF4 suppressions of proliferation and angiogenesis in endothelial and vascular smooth muscle cells. Biochem Biophys Res Commun. 2013;437:625-631. 29.

Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to subarachnoid

30.

TE D

hemorrhage visualized by computerized tomographic scanning. Neurosurgery. 1980;6:1-9. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25:402-408. Vikman P, Beg S, Khurana TS, Hansen-Schwartz J, Edvinsson L. Gene expression and

EP

31.

molecular changes in cerebral arteries following subarachnoid hemorrhage in the rat. J

32.

AC C

Neurosurg. 2006;105:438-444. Zheng B, Han M, Wen JK. Role of Kruppel-like factor 4 in phenotypic switching and

proliferation of vascular smooth muscle cells. IUBMB Life. 2010;62:132-139.

33.

Yoshida T, Kaestner KH, Owens GK. Conditional deletion of Kruppel-like factor 4 delays downregulation of smooth muscle cell differentiation markers but accelerates neointimal formation following vascular injury. Circ Res. 2008;102:1548-1557.

34.

Selvamani A, Williams MH, Miranda RC, Sohrabji F. Circulating miRNA profiles provide a

R20

Kikkawa et al.

ACCEPTED MANUSCRIPT

biomarker for severity of stroke outcomes associated with age and sex in a rat model. Clin Sci (Lond). 2014;127:77-89. 35.

Wu J, Du K, Lu X. Elevated expressions of serum miR-15a, miR-16, and miR-17-5p are

36.

RI PT

associated with acute ischemic stroke. Int J Clin Exp Med. 2015;8:21071-21079.

Mayr M, Zampetaki A, Kiechl S. MicroRNA biomarkers for failing hearts? Eur Heart J. 2013;34:2782-2783.

Nakagomi T, Kassell NF, Sasaki T, Lehman RM, Hongo K, Ogawa H, et al. Time course of

SC

37.

the blood-arterial wall barrier disruption following experimental subarachnoid haemorrhage.

Sasaki T, Kassell NF, Zuccarello M, Nakagomi T, Fijiwara S, Colohan AR, et al. Barrier disruption in the major cerebral arteries during the acute stage after experimental

EP

TE D

subarachnoid hemorrhage. Neurosurgery. 1986;19:177-184.

AC C

38.

M AN U

Acta Neurochir (Wien). 1989;98:176-183.

R21

Kikkawa et al.

ACCEPTED MANUSCRIPT

Figure legends Fig. 1. Timeline of sampling and investigation flowchart. (A) Timeline of cerebrospinal fluid

RI PT

(CSF) and blood sampling after subarachnoid hemorrhage. (B) Flowchart of gene expression analysis of miRNAs and KLF4.

SC

Fig. 2. Time course of changes in microRNA-15a (miR-15a) expression following subarachnoid hemorrhage (SAH). Quantitative real-time PCR analysis of exosomal miR-15a in the plasma (A)

M AN U

and CSF (B) following SAH. Data are the means ± SEM (n = 4-6 per time point). The expression level seen on day 0 (control) was assigned a value of 1.0. *P < 0.05 versus day 0.

Fig. 3. Time course of changes in mRNA expression of Kruppel-like factor 4 (KLF4) following SAH. Quantitative real-time PCR analysis of KLF4 in the plasma (A) and CSF (B) following SAH.

TE D

Data are the means ± SEM (n = 4-6 per time point). The expression level seen on day 0 (control) was assigned a value of 1.0. *P < 0.05 versus day 0.

EP

Supplementary Fig. S1. Time course of changes in expression of microRNA-15a (miR-15a) and microRNA-6724 (miR-6724) following subarachnoid hemorrhage (SAH). Quantitative real-time

AC C

PCR analysis of miR-15a (A, plasma; B, CSF) and miR-6724 (C, plasma; D, CSF) following SAH. Data are the means ± SEM (n = 6 per time point). The expression level seen on day 1 was assigned a value of 1.0. *P < 0.05 versus day 1.

R22

ACCEPTED MANUSCRIPT

Table 1. microRNAs that were up- or down-regulated in both CSF and plasma in two samples on day 3 after subarachnoid hemorrhage. CSF 1

CSF 2

Plasma 1

Plasma 2

4.60

3.76

-4.45

-1.84

hsa-miR-19b-3p

4.00

4.05

-3.61

-2.01

hsa-miR-15a-5p

3.66

2.31

-4.56

hsa-miR-103a-3p

4.00

2.29

-3.05

hsa-miR-15b-5p

3.34

2.36

-3.88

hsa-miR-92b-3p

4.03

2.31

-2.75

hsa-miR-29a-3p

1.36

2.29

-1.44

hsa-miR-191-5p

1.55

2.20

hsa-miR-22-3p

2.17

1.80

hsa-miR-30d-5p

1.23

1.41

hsa-miR-6724-5p

-1.54

hsa-miR-2467-3p

1.72

EP AC C

-1.69 -1.57 -2.25 -1.57

SC

-1.17 -1.28

-2.82

-1.12

-2.02

-1.32

-1.33

-2.33

-1.70

1.84

1.20

1.87

M AN U

-2.88

TE D

Abbreviations: CSF = cerebrospinal fluid

RI PT

hsa-miR-16-5p

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights

1. Microarray analysis of CSF and plasma of SAH patients was performed and differentially expressed miRNAs were identified.

following SAH was demonstrated.

RI PT

2. The time course of changes in the expression of miR-15a and KLF4 in the CSF and plasma

3. Early and persistent decrease in KLF4 followed by an increase in miR-15a may contribute to

AC C

EP

TE D

M AN U

SC

vascular proliferation or angiogenesis after SAH.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Supplementary Table S1. S1. microRNAs microRNAs that were upup- or downdown-regulated in CSF or plasma in two samples on day 3 after subarachnoid hemorrhage. Up- or down-regulated

Up- or down-regulated

microRNA in CSF

microRNA in plasma

CSF 1

Plasma 1

4.60

3.76 hsa-miR-3194-3p

1.17

1.02

hsa-miR-19b-3p

4.00

4.05 hsa-miR-4740-5p

1.39

1.14

hsa-miR-92a-3p

4.42

3.10 hsa-miR-2467-3p

1.20

1.87

hsa-miR-486-5p

5.98

4.31 hsa-miR-5008-5p

1.34

1.19

hsa-miR-15a-5p

3.66

2.31 hsa-miR-19b-3p

-3.61

-2.01

hsa-miR-103a-3p

4.00

2.29 hsa-miR-15b-5p

-3.88

-2.25

hsa-miR-15b-5p

3.34

2.36 hsa-miR-144-3p

-2.81

-2.09

hsa-miR-106b-5p

3.07

2.32 hsa-miR-6716-5p

-2.58

-2.25

hsa-miR-92b-3p

4.03

2.31 hsa-miR-15a-5p

-4.56

-1.69

hsa-let-7b-5p

2.76

2.94 hsa-miR-16-5p

-4.45

-1.84

hsa-miR-100-5p

2.01

2.50 hsa-miR-103a-3p

-3.05

-1.57

hsa-miR-4497

4.22

1.28 hsa-miR-106a-5p

-3.37

-1.59

hsa-miR-204-3p

4.79

1.74 hsa-miR-451a

-4.52

-1.68

hsa-miR-22-3p

2.17

1.80 hsa-miR-22-3p

-2.82

-1.12

2.43

1.91 hsa-miR-30d-5p

-2.02

-1.32

2.93

1.36 hsa-miR-191-5p

-2.88

-1.28

1.36

2.29 hsa-miR-92b-3p

-2.75

-1.57

1.55

2.20 hsa-miR-6724-5p

-2.33

-1.70

1.62

1.53 hsa-miR-29a-3p

-1.44

-1.17

1.93

1.51

1.39

1.49

hsa-miR-30d-5p

1.23

1.41

hsa-miR-614

1.76

1.66

hsa-miR-3131

1.19

1.06

hsa-miR-2467-3p

1.72

1.84

hsa-miR-1233-5p

1.71

1.59

hsa-miR-6840-3p

1.02

1.12

hsa-miR-204-5p

-3.83

-1.95

hsa-miR-1260b

-2.05

-1.10

hsa-miR-187-5p

-1.55

-1.39

hsa-miR-29a-3p hsa-miR-191-5p hsa-let-7c-5p hsa-let-7d-5p

AC C

hsa-miR-99a-5p

M AN U

TE D

hsa-miR-1288-3p

EP

hsa-miR-140-3p

SC

hsa-miR-16-5p

Plasma 2

RI PT

CSF 2

-1.40

-1.47

hsa-miR-1915-3p

-1.04

-1.38

hsa-miR-4488

-1.42

-1.31

hsa-miR-4787-3p

-1.13

-1.08

hsa-miR-5100

-1.87

-1.53

hsa-miR-6724-5p

-1.54

-1.33

hsa-miR-6784-5p

-1.18

-1.02

hsa-miR-6803-5p

-1.07

-1.55

hsa-miR-6836-3p

-1.03

-1.42

hsa-miR-6869-5p

-1.52

-1.42

AC C

EP

TE D

M AN U

Abbreviations: CSF = cerebrospinal fluid

SC

hsa-miR-1268a

RI PT

ACCEPTED MANUSCRIPT